BimmerFest BMW Forum banner

335d Torque/BT data collection

38K views 86 replies 12 participants last post by  TDIwyse 
#1 ·
Seems like there was some interest is this type of thing from the other thread. So if there's interest, and you like data, please contribute.

I've just started playing with some additional fields in the Torque app (not all of the fields are able to get data) that I hadn't looked at previously. There's several others I'd like to view and record as well in following posts (like comparing how much fuel is used at idle when in D or N, or with the AC on or off, etc.)

But here's some of the data from this mornings commute. It was freezing fog conditions so no wind and the drive was slower than normal (~49-54 mph most of the way) and I hit 3 of the 5 stop lights. Better than normal average mpg for the trip (due to lower speeds and no wind). Found the Air/Fuel ratio and Volumetric Efficiency calculation interesting.

Need to read more about the HP/TQ readings as they only show up sporadically (not sure if it is triggered on G force, so at steady state it doesn't record data). Anyway, I zoomed in on a section where I accelerated from a stop light (at restrained levels due to the ice). These two numbers look like fun things to experiment with . . .
 

Attachments

See less See more
1
#42 · (Edited)
The JBD throws codes because the ECU is being sent a signal that there is not enough rail pressure and hence a fuel pump problem (that is the code). Which makes sense knowing how the JBD operates. As one dials up the JBD, the ECU get a signal that even less fuel is being delivered which in-turn the ECU tells the fuel pump to pump more fuel to compensate. When limp mode/SES occurs in certain instances when the JBD is set at 100% or close to it a fail-safe is triggered to reduce power to the engine in response to what the ECU deems is likely a fuel pump failure.
 
#43 ·
I'd love to have access to someone's car who has a BT and the Evolve or RENNtech tunes to compare the rail pressures, injector timing/openings, EGT's, etc.

Any ideas on what limits the remap's pwr levels? Is it also rail pressure related due to flowing more fuel than the HPFP can feed?
 
#44 ·
I'd love to have access to someone's car who has a BT and the Evolve or RENNtech tunes to compare the rail pressures, injector timing/openings, EGT's, etc.

Any ideas on what limits the remap's pwr levels? Is it also rail pressure related due to flowing more fuel than the HPFP can feed?
Most of these tunes produce around the same amount of power (hp/torque) depending on which dyno plot you believe.

Regardless of the tune, most of the power is generated by adding more fuel, although some tunes claim to increase boost pressure too (Renntech, Evolve etc), it tends to be only a slight increase in this area. How the fuel is increased/delivered is what likely differentiates some of the tunes vs the JBD.

Many diesel tunes claim to remap the fuel injection timing, hence they increase the amount of fuel by increasing the number of times the fuel pump injects fuel under certain load conditions rather than just increasing fuel pump pressure across the board.
 
#45 ·
My other common rail diesel's (Cummins, Liberty CRD) have used methods where they increase the duration of the main pulse (along with increase injection timing) as the main method to get more fuel. I actually prefer this method to the "box" type rail pressure only methods as it's substantially less stress on the injector bodies and HPFP. But you still get to a point where the HPFP can't maintain adequate pressure in the rail due to flow constraints. Hopefully that's not the common limitation in the JBD/Evolve/RENNtech setups.

The exhaust temps and DPF restrictions don't seem to be limits from the data I'm seeing. Although as the DPF ages it might add more restriction. Also, in the X5 if you're towing into headwinds or going up mountains the EGT's would become substantially higher due to prolonged high demand situations.

The intercooler seems like it could stand to be improved a bit as the the intake temps are increasing ~ >30C in a matter seconds.
 
#46 ·
My other common rail diesel's (Cummins, Liberty CRD) have used methods where they increase the duration of the main pulse (along with increase injection timing) as the main method to get more fuel. I actually prefer this method to the "box" type rail pressure only methods as it's substantially less stress on the injector bodies and HPFP. But you still get to a point where the HPFP can't maintain adequate pressure in the rail due to flow constraints. Hopefully that's not the common limitation in the JBD/Evolve/RENNtech setups.

The exhaust temps and DPF restrictions don't seem to be limits from the data I'm seeing. Although as the DPF ages it might add more restriction. Also, in the X5 if you're towing into headwinds or going up mountains the EGT's would become substantially higher due to prolonged high demand situations.

The intercooler seems like it could stand to be improved a bit as the the intake temps are increasing ~ >30C in a matter seconds.
All of these tunes require the fuel pump to perform double duty; now weather or not one method shortens the life of the pump anymore than the other is anyone's guess. The tuners claim their method is "safer" while the fuel box camp claim the tuners are full of it. I am of the mind-set that it is 6 of one, half a donzen of the other. That being said, I think anytime the ECU is in control vs being tricked into doing something else would be the preferred method.

The DPF has shown to be little if any restriction as evidenced by an E90Post member that removed his DPF and did not see any hp gain. In fact he reinstalled the DPF because he did not like the increase in soot. This also seems to be supported by an artcile I posted in my exhaust thread on cat back exhaust systems on turbo diesels.

I agree an IC would be a nice add, esp for those that have a tune. Anytime power is increased logic would dictate that temps increase, so one would think there would be a benefit to adding a more efficient IC, not much unlike the benefits the 335i owners see.
 
#50 ·
One of the plots didn't upload . . . this is the time of the main injection pulse * the injection angel (start of injection). Kind of a way to show the impact of increased duration and timing. Since the JBD fools the ECM into making more pressure, for those situations where the JBD is utilized, for the same injector opening duration, more fuel goes in due to the rail being at higher pressures.
 

Attachments

#57 ·
So being new to diesel tuning, I'm assuming injection angle references crank position and by beginning injection earlier is not only adding fuel by increasing duty cycle but, it is also similar to adding ignition timing in a gasoline motor, right?

My next question is do we know if the ECU has some type of detonation protection that retards injection timing like a gas car would pull ignition timing if something were to see ridiculously high loads or something go wrong?
 
#51 ·
Well, needed to run some errands so I stopped off and downloaded some data.

This is one of the best of the remap only runs (pwr results varied from ~224-230 on multiple runs) compared to the single runs of the remap w JBD at 0% and 65%.
 

Attachments

#52 ·
A couple more pulls were performed after getting some local fuel back in the tank (~1/2 tank added after long drive the previous night). Think the other fuel was less than optimum based on these results.

Temp was more than 10F warmer than previous testing. Below shows comparisons of the remap stacked with JBD at 0 and ~65%. The "Main" and Run 1 are 0%, Runs 2/3 are 65%. Again, was being careful not to give full fuel at low rpm's . . .

Also showing some of the "speed" data which is interesting. For these 3rd gear pulls the full fueling is occuring between ~45-75mph. The 50-70 times are thus right in the main thrust of the test. The reported times are nice.

Also, still not seeing any engine parameters that are over stressed. Although the pulls happen so fast that this is somewhat expected. If you were continuing this kind of stress (towing or racing for example) it would likely be a much different story.
 

Attachments

#54 · (Edited)
Thanks for gathering all this data. To avoid making potentially wrong assumptions on this chart which run is which? Also, I didn't see the info in there and maybe you don't want to disclose it out of courtesy to Evolve but, would you mind throwing up a plot of stock boost and remap boost vs. RPM? If not that's cool too but, could you at least clarify if it is a significant increase in boost or only a few PSI? I'm not sure if using that low of a gear will even give accurate boost numbers but, I just like to see data.
 
#55 ·
The owner has had some interaction with Evolve and they haven***8217;t said to not share any of this info. They, by the way, have been good to work with. Same for VS. And to be clear, they do not recommend stacking the JBD on top of their tune.

The boost info is back here: http://www.bimmerfest.com/forums/showpost.php?p=6870998&postcount=49

The JBD run is the same boost as stock. It does not touch boost (one of the previous posts has stock/JBD and verifies that). These numbers are absolute pressure (normal atmosphere + boost). Converting hPa to PSI and subtracting 1 atmosphere brings the numbers to ~ 26psi stock and 30psi remap.

The BT data is on a different computer and it can be plotted vs rpm. Will try to do that tonight. But the boost response is unbelievably fast and occurs right away even at low rpm***8217;s. Most likely due to how BMW has executed the twin turbo setup. Just lovely.

I agree the naming of the Gtech data has been poor. When the Gtech gets powered up it defaults to Jan 17 2017 and the time seems to be random. And I haven***8217;t been good at going in and re-formating the dates/times/conditions etc . . . I tried to correct that starting on post 52 which shows multiple runs with the remap and JBD stack. The plot you showed is Run1 = Evolve (2nd best of multiple runs), Run 2 Evolve+JBD 0% (only 1 run on this combo that day), Run 3 Evolve + JBD 65% (only 1 run on this combo and the last run).

A note on the Gtech data. It includes more losses than a 2 wheel dyno: 4 wheels are turning instead of 2 (more rolling resistance), wind losses (~20 hp at top end of run based on Cd and frontal area), and the pulls are in 3rd gear so there's more drivetrain loss than a ~1:1 gear ratio). And it's subject to variations due to road surface roughness, flatness, the accuracy of the weight, etc . . . But it's been very usefull over the years for testing relative impacts of mods.

From the Gtech data that***8217;s been gathered it appears the remap is acting similar to the JBD at ~65% but with better EGT***8217;s and IAT***8217;s and lower peak rail pressures. The remap with JBD at 0% is similar to the JBD at the max before it throws a limp mode.

Oh, and the owners reported hand calculated mileage for the last two fillups with the remap covering ~300+ mls, including all these full fueling runs, has averaged 39.2 mpg. This is at least 5% better than pre map for similar driving conditions.
 
#56 ·
Here's an example of the remap boost response. The BT sample rate is only about once per second. The "torque" value is from the ECU "Internally Calculated Torque". This is included to show when the full fueling hits. So somewhere between 1787 and 2251 rpm's (that's the limit on the time step resolution for the BT data) the boost is already at its max value.

If Evolve would like all this info removed it will be done. However this type of data would seem to be valuable for an informed decision . . .
 

Attachments

#59 ·
Some interesting data collected today on a local 335d with an Evolve remap stacked with JBD on 0 and post IC/pre IAT water/methanol injection. Temps were 98F and ~41% Relative Humidity. Setup utilized a 200psi pump, 550ml/min nozzle, -20F washer fluid (~30% Methanol/70% water), and a Snow 0-5V MAF based controller fed with the MAP sensor. The Map sensor puts out ~1-5V based on intake pressure so the injection response is boost-based, even though it‘s using a MAF controller.

Very significant impact to intake temps during full fuel runs, as well as significantly reduced temp recovery time post runs. Also good impact to HP based on OBD acceleration data and Gtech results (will try to post Gtech plots tomorrow). Max EGT’s measured were 544C with and 537F without injection. Still well below temps seen during DPF regens.
 

Attachments

#60 ·
Here's the plots. Was careful at low rpm's and didn't start full fuel until just above 2k on the rpm's. Kind of neat how around the same rpm point that the IAT's in the plots above start to diverge (injection starts), so too the HP diverges. Not a substantially higher amount of HP at these settings, but a lot more area under the curve.

From the looks of the recorded data at these hot conditions, and the safe readings on the EGT's, DPF pressures, etc. it looks like there's more room. . .
 

Attachments

#62 ·
There was time spent today fine tuning the controller settings. Turns out the MAP sensor voltage output range is peaking at ~3.3V with the remap. So the controller was adjusted for its "max" setting to be closer to this value instead of the 4.5V it was initially set at. Also increased the "start" point to closer to ~2.6V so it doesn't inject unless the vehicle is really being pushed. This may be reduced a bit pending further data.

Attached is a comparison of the "new" vs. "initial" settings in regards to post intercooler "boosted air intake" temps and EGT's for the same Remap/JBD 0% stack. Care was again exercised to avoid full fueling at low rpm's.

The 90F temps today were not quite as hot as the 98F day's data (~4.5 deg C lower) but the impact to IAT's and EGT's is quantifiably substantial with the additional injection at higher boost levels. In fact the IAT's with the new setting are seen to actually be reduced to just below the starting value during the pull, and are kept to within 2C of the starting temps at 4200 rpm where the peak hp of the engine is typically found. Also the EGT's are substantially lowered at this rpm point from 503 to 477 ( 937.4 F to 890.6 F).

The owner also put a switch in play so the injection can be disabled during DPF regens.

On a separate note, looking into ways to minimize low rpm torque but extend high rpm torque of the stack. Think there might be a way. It would involve making an RPM variable resistor to use instead of the manual potentiometer on the JBD. It would involve using a Freq to Voltage converter (something like this: https://www.national.com/ds/LM/LM2907.pdf) where one takes the tach signal (lower right pin on the OBD port has a beautiful tach signal) converting this to voltage that adjusts the variable resistor in the JBD (something like this: http://www.vishay.com/docs/70598/70598.pdf). And all of a sudden one gets the ability to add in the JBD rail pressure increase that is a function of rpm which would reduce the problem of too much low end torque. The circuit to be setup so at rpm's below ~3k nothing above JBD at 0% is utilized, and then from 3k to ~4.5k the JBD distortion increases to the final desired level. The distortion adds to the remap only above ~3k and at increasing values to compensate for the torque roll-off. Now we have a torque curve that is essentially flat out to nearly 4.5k and could stay at/below the ~700Nm/520 lb-ft rating of the tranny.

Also including some pics of the place in the pipe the probe is located and how it fits in the chassis. It's tight in there . . .
 

Attachments

#63 ·
TDIwyse,

I am actually doing what you are talking about now and am installing my water meth kit starting this weekend. What I did is used an AEM FIC to intercept the TPS, Rail pressure, Boost and MAF sensors, I have RPM as well. I can control the water meth pump and nozzle solenoid using RPM and boost pressure. I understand your concerns with too much torque in the lower rpms. It’s also my goal to limit the torque down low but get as much power/torque as I can in the upper RPM. Using the AEM, I have been able to fine tune the fuel pressure to push the power as much as I can for all RPM and loads while not getting knocks or any codes. FYI, max rail pressure you can push is 1800bar or 4.5v from the rail sensor. Anything over that and the rail bypass valve returns it to the tank. Great work!!!


-Ed
 
#64 ·
TDIwyse,

I am actually doing what you are talking about now and am installing my water meth kit starting this weekend. What I did is used an AEM FIC to intercept the TPS, Rail pressure, Boost and MAF sensors, I have RPM as well. I can control the water meth pump and nozzle solenoid using RPM and boost pressure. I understand your concerns with too much torque in the lower rpms. It's also my goal to limit the torque down low but get as much power/torque as I can in the upper RPM. Using the AEM, I have been able to fine tune the fuel pressure to push the power as much as I can for all RPM and loads while not getting knocks or any codes. FYI, max rail pressure you can push is 1800bar or 4.5v from the rail sensor. Anything over that and the rail bypass valve returns it to the tank. Great work!!!

-Ed
Thanks.

And I really like your approach with the AEM FIC and enjoyed your thread. Also appreciate you sharing results with the community. Please keep sharing!

The OBD pin 9 tach output to voltage circuit is made and verified its performance (Voltage out to RPM). Hoping to interface with the JBD this afternoon. It will linearly increase the JBD from 0 to 70% from 2500 rpm to 4100 rpm. Depending upon measured data the top end can be increased further.
 
#66 ·
Comparison of Evolve/7030H2O/progressive JBD with post #60 data with JBD fixed at 0%. The progressive JBD mod data looked good so the top end was increased. This run was with the top end set to ~90% on the total JBD capability. Also, this new data was on a heat soaked vehicle with multiple full fuel runs on it in ~91F humid ambient conditions. The owner wasn't being very cautious with low rpm full applications for the new runs to test the low rpm torque curve characteristics.

The mod definitely helps extend the top end torque/hp and gives more area under the curve. Also helps minimizes peak low rpm torque for non-careful applications of fuel (see post #27 for example where normal peak torque for 3rd gear pulls occurred ~2100 rpm ).
 

Attachments

#67 ·
Some interesting data regarding pre/post EGR recal data for differences in EGR Duty Cycle and exhaust gas temps for my drive to work. There's a big difference in these characteristics. Not sure I like the extra EGR utilization to increase average DPF temps. Thought perhaps the recall added some post injection to increase DPF temps, but it appears it's just more EGR . . .
 

Attachments

#68 · (Edited)
As you suggest the elevated temperatures may not be only from EGR. Looks like the EGR is being operated more like an on/off device instead of modulating. This would be one way of reducing the soot build up as I would expect build up to occur more during the partial opening than full open or full closed.

The elevated temperatures may be by design to reduce fouling of the intake from EGR, not so much for the DPF. I suspect the eleveted temps are from late power cycle fuel addition, which could be a cause of the reduced economy some are experiencing.

Regardless, very interesting for us engineer types.
 
#69 ·
As you suggest the elevated temperatures may not be only from EGR. Looks like the EGR is being operated more like an on/off device instead of modulating. This would be one way of reducing the soot build up as I would expect build up to occur more during the partial opening than full open or full closed.

The elevated temperatures may be by design to reduce fouling of the intake from EGR, not so much for the DPF. I suspect the eleveted temps are from late power cycle fuel addition, which could be a cause of the reduced economy some are experiencing.

Regardless, very interesting for us engineer types.
The EGR goes to 0 under certain conditions like costing in gear or full fuel situations or during DPF regen events. The data points with 0 EGR Duty cycle in the above plots are for coasting are heavy acceleration situations.

The 2 "Post Injection" events have the same behavior pre and post recal. Once the coolant temp hits 140F they end, except for DPF regen events when they are used to maintain DPF temps sufficient to regen.

The only major differences I see in the data pre/post recal is heavier utilization of EGR and slightly increased DPF temps. I really wished they didn't increase the EGR utilization. That's a major cause of intake carbon buildup.
 

Attachments

#73 ·
When I had my recall performed, I got the SA to print a copy of the Technical Service bulletin for me (not sure if he was supposed to, so no names to protect the innocent). This quote might help the discussion:

"Under certain driving conditions, the High-pressure (HP) EGR valve may become restricted with excessive soot. This restriction impedes the emissions performance of the HP EGR valve."

and

"An optimized EGR valve has been develped to limit this soot buildup."

Interesting that what BMW wrote seems to be contradicted by what TDIwyse is finding.
 
#74 ·
Good stuff Joseph; first time I've seen something from BMW that specifically states the reason for ther recall. Although I find it interesting they only mention soot build-up as it relates to the EGR valve. :confused:

I wonder what impact this partially restricted EGR vavle had on soot build-up in the combustion chamber and more specifically the build-up that some members stated lead to the replacement of the heads in several cars.
 
#75 ·
I wonder what impact this partially restricted EGR vavle had on soot build-up in the combustion chamber and more specifically the build-up that some members stated lead to the replacement of the heads in several cars.
I had assumed everyone was aware the soot was the root of the recall, my SA had read me the details of the TB. And they way I figure it, it was those who drove a lot and accumulated lots of mileage quickly, ending up needing a head replacement are the ones that triggered the redesign/recall. I believe the soot just plugged up everything (EGR, Intake and heads). So we can thank these road warriers for preventing the problem from occuring for the rest of us, although at an apparent loss of fuel economy.

I asked the SA about walnut blasting the heads to clean them. He said while that would work, it was still less expensive just to replace, rather than clean the head and intake.
 
#77 ·
I'm sorry to do this, but going back and looking at some other data shows something that complicates the EGR data I posted yesterday . . .

The attached is from Feb of this year (pre recall) on a very long road trip. It was recorded with the Android Torque app at 5 sec intervals.

The EGR behavior appears to vary with the DPF loading. The data for the drive started with a fairly fresh DPF condition. The data shows the car doing a DPF regen around the "1700" time points where the EGR is very low for ~10 mins. Note that the closer the car gets to DPF regen time the lower the EGR utilization. And after the regen the EGR rate increases to heavy utilization. I also included the "engine load" and "speed" parameters to show the similarity in operating conditions as those things can also impact EGR.

So with the previous data I posted I'm not sure of the DPF condition for the "pre recal" EGR example for that mornings drive to work. The "post recal" behavior is on a recently regenerated DPF which means it's similar to the behavior in this graph.

So basically I'm retracting the conclusion I made that the EGR behavior is significantly different.

Sorry for the confusion . . .
 

Attachments

#79 ·
Spent some time this weekend with the GTech device to analyze launching traction issues that seemed potentially helpful to others. Also, there was some additional tweaking on the above car this weekend. The progressive JBD mod was further tweaked (to reduce low rpm distortion to below "0" level to minimize peak low rpm torque, and push the distortion further into the rpm and take the peak distortion to the ~100% level to further enhance top end power) as well as running 50/50 water/methanol (recently added an AEM flow meter and the car is sucking ~650ml/min at peak fueling).

The GTech device shows some very interesting effects on the launching procedures. Below is 4 seperate runs with various brake boosted launching rpm's and fueling procedures. Additonally two runs had the rear street tires at 41psi and two at 33psi. 3 had DCT on, 1 had all traction control off.

It's neat to see the impact of these variables on the G forces and speeds. Obviously the lower psi on the tires helps with traction, and it also appears this car needs to not be floored immediately as the tires can't handle the traction at the two psi settings that were examined. Further reduction in psi, or using drag radials should improve the traction to a point where more fuel can be applied sooner for better 60' launches.

Also thought the Black vs Green run was interesting. The green had a better 60' time, but because the tires lost traction, and no traction control algorithms were used to cut back pwr, the 330' time was actually worse.

Also, used the Black run (3rd gear portion) to compare to the earlier recorded hp/tq curves using the previous JBD/H2O injection setup. The torque curve is pushed further into the rpm range and provides for nearly 20more hp from 3400-4400 rpm. It's probably not quite that much as the temps were a lot cooler this weekend, although it was into a 10mph headwind . . .
 

Attachments

#81 ·
Hmmm, perhaps I can clear up some potential area's where the data, or presentation of the data, wasn't clear.

Important points: The best comparisons of various combo's will come from looking at the data collected on the same day with the same fuel and at the same ambient temps on the same roads with the same winds and same DPF loading and same heat soaked engine conditions (before the H2O/methanol injection this aspect had more effect and could skew results ~10hp). That's a lot of different variables right there even before looking at the different power adder combo's. And there's a lot of different combo's in this thread.

So, the post with the first quote above (page 2 post #27) was with 70F and no wind, stock vehicle with the JBD at 0, ~65 and ~85%. Those results were ~214, 237, 253 hp and 363, 378, 402 ft-lb. No comparison without the JBD.

Post #40/41 shows a comparison of stock vs JBD at 0 and 65%. These were done back to back, so relative to each other they should be pretty close, but not necessarily a great comparison to the post #27 data. This comparison showed ~209, 221, 241hp and 351, 368, 381 lb-ft (slightly more than the #27 data for the JBD at 0 and 65). So the JBD at 0 seemed to add ~12 hp above stock and the JBD at 65% added ~32hp with the GTech hp/tq measurement on this day.

Post #49 starts with the Evolve evaluation. There's several caveat's given for why direct comparisons's to previous data needs to be considered. The general summation I gave was this: "But here’s the summation of the hp/tq #’s. The JBD @ ~85% produced similar TQ #’s as the remap, but a bit better HP. Now it might be that the tank of Illinois fuel in the vehicle has lower cetane or BTU’s than what was in the vehicle for the previous tests. Don’t have a way to test that. "

On this same post it's noted that the stack of the Evolve remap with the JBD at 65% produced ~30hp increase over the remap alone. The curves on post #51 (the tq curves for the remap/JBD stack are not accurate due to not giving full fuel at low rpm's to avoid excessive low rpm torque) show ~ 33.5 hp increase for ~65% and just a few hp for the 0%. However, the remap only data was from a different day/fuel/temp/etc, so an absolute comparison isn't valid. However, what is valid to say is the delta increase of the remap with the JBD follows a very similar trend as that from post #41.

The last quote you put up is from post #52 and includes #'s from post #51 as well as #'s from a different day/fuel/environment condition. The new data showed ~19hp increase from the 0-65% JBD setting instead of the ~30hp from previous data groups. However it was a warmer day and heat soak could've been involved in skewing those #'s for the 65% test, which was performed a few mins after the 0% run.

Clear as mud?
 
This is an older thread, you may not receive a response, and could be reviving an old thread. Please consider creating a new thread.
Top